Title :
The flow amounts prediction of BFG with an improved PSO-BP algorithm
Author :
Junpeng Li ; Changchun Hua ; Yinggan Tang ; Xinping Guan
Author_Institution :
Coll. of Electr. Eng., Yanshan Univ., Qinhuangdao, China
Abstract :
The byproduct gas of blast furnace is one of the most significant energy resources of an enterprise. Because of the complex chemical reaction in blast furnace, the trend of the flow amounts of BFG is hard to be predicted. In this paper, we proposed an improved PSO-BP algorithm which has good performance compared with PSO-BP algorithm and BP algorithm. The prediction accuracy and running time of our algorithm can meet the demands of industry. The results provide guidance for industrial operations.
Keywords :
backpropagation; blast furnaces; neural nets; particle swarm optimisation; production engineering computing; BFG flow amounts prediction; blast furnace byproduct gas; complex chemical reaction; improved PSO-BP algorithm; Artificial neural networks; Blast furnaces; Convergence; Correlation; Prediction algorithms; Predictive models; BP; Blast furnace gas (BFG); PSO; grey correlation; prediction;
Conference_Titel :
Control Conference (CCC), 2013 32nd Chinese
